AI in multi-omics analysis in AMR

Antimicrobial resistance (AMR) is a global public health threat that has severely jeopardised years of progress in controlling infectious diseases. The last few years have witnessed a dramatic expansion in the high-throughput "-omics" technologies, such as genomics, transcriptomics, proteomics, metabolomics, etc. The multi-omics approaches have established that AMR develops due to complex interactions among various biomolecules, including genetic variations, regulatory networks, differential protein expressions, metabolic adaptations, and environmental pressure. But the scale, heterogeneity and high dimensionality of data generated via multi-omics analyses far exceed the capacity of traditional statistical and rule-based analytical methods. Thus, Artificial Intelligence (AI) is needed for extracting meaningful patterns from the complex high-dimensional biological data generated during multi-omics analyses. Artificial intelligence (AI), encompassing machine learning and deep learning techniques, is a robust way to combine different types of -omics data, model non-linear dependencies, and get predictive and mechanistic insights into AMR. This chapter provides a comprehensive and critical overview of the role of AI in multi-omics analysis of AMR, conceptual foundations, methodological advances, representative applications and translational implications. Additionally, significant applications of AI-driven multi-omics AMR research are discussed. Finally, limitations of this strategy and future directions are examined. To fully realise the potential of AI-driven multi-omics strategies for combating AMR, it will be necessary to address their limitations through standardised data generation, robust validation, and interdisciplinary collaboration.